Anthropic Posts $80.7 Billion Net Loss as Revenue Jumps 1,300%
Anthropic reported a net loss of $80.7 billion in the first half of this year, almost twice its total loss for all of last year. Most of the loss came from changes in the estimated value of financial instruments, rather than actual cash spending. The company has also agreed to spend as much as $750 billion on computing power and infrastructure. Meanwhile, trading in insurance against debt defaults by major technology companies has increased 4.5 times.
Anthropic reported a net loss of more than $80 billion in the first half of this year, according to a filing with securities regulators. The report comes ahead of the company's planned initial public offering (IPO) next month, when it plans to sell shares to the public for the first time. The loss reflects the huge amounts of money spent building the infrastructure needed for artificial intelligence (AI). However, concerns remain about whether AI will earn enough money to justify these investments.
Up to $750 Billion for Computing Power
Anthropic's revenue reached $16.3 billion in the first half of this year, an increase of more than 1,300% from $1.1 billion during the same period last year. The figures came from the company's S-1 registration statement, which The Wall Street Journal obtained on the 20th. Despite this rapid growth in revenue, Anthropic reported a net loss of $80.7 billion over the same six months. That was almost twice the $42 billion loss it reported for the whole of last year.
Spending on infrastructure was a major reason for the growing loss. Anthropic spent $9 billion on computing power and infrastructure in just the first half of this year. That was already more than the $7.3 billion it spent during all of last year. The Journal reported that the company has signed agreements to obtain future computing resources worth $450 billion. Most of these agreements, the report said, cannot be ended early. If ongoing negotiations are also included, Anthropic could need to commit as much as $750 billion just to secure enough computing power.
However, most of the $80.7 billion loss in the first half did not come from money actually being spent. A total of $72 billion came from changes in the estimated value of convertible financial instruments. These include convertible bonds, which the company issued to raise money for infrastructure investments. The value of these instruments is linked to the company's share price. As a result, the related charges are designed to increase as Anthropic's estimated value rises after it lists its shares next month.
Trading in 'Default Insurance' for Big Tech Rises 4.5 Times
Questions about whether AI can make enough money continue to grow. Concerns about the risk of companies failing to repay their debts first emerged around CoreWeave, a major provider of cloud computing services for AI. These concerns are now spreading to large technology companies. The Financial Times reported on the same day, citing a recent Morgan Stanley report, that weekly trading in credit default swaps linked to bonds issued by Oracle, SpaceX, Alphabet, Amazon, Meta, Broadcom and Nvidia averaged $9 billion last month. This was about 4.5 times the level in July, two months earlier, when trading was below $2 billion a week.
A credit default swap (CDS) works like insurance. Buyers pay for protection in case a company fails to repay its debt. Trading in these contracts has increased as investment in AI infrastructure has grown. The rise in this type of debt insurance suggests that companies are raising money for high-risk investments at an even faster pace. The Financial Times noted that a rush to borrow money to pay for trillions of dollars in AI infrastructure is increasing demand for financial products that work like insurance.
Morgan Stanley reported that companies have issued $466 billion in bonds so far this year to pay for AI-related infrastructure, including data centers and semiconductors. That is more than four times the $101 billion issued during the same period last year.
Morgan Stanley also predicted that major technology companies focused on AI would have a record amount of negative free cash flow next year, reaching about $540 billion when lease costs are included. Free cash flow is the money a company has left from its operating cash flow after paying for capital investments, such as buildings and equipment.
Originally reported by Cho Yang-joon for Seoul Economic Daily.
Translated from Korean using AI. Quotes from foreign sources are based on Korean-language reports and may not match the exact wording of the original statements.